机器学习---学习记录

day one–吴恩达机器学习课时1-课时7

1. 监督学习（supervised learning）: right answer given, input and output has a certain relationship. it includes regression problem(predict continuous valued output)—house price , classification problem(discrete valued output) --tumber 0 or 1
2. 无监督学习（unsupervised learning）: result unknown, no label data, include clustering algorithms, Cocktail party algorithm
3. 监督学习模型描述：
   训练集（training set）
   notation:
   m–number of training examples
   x–input variable/features
   y–output variable/target variable
   (x,y)–one training example
   (x^(i) ,y^(i))–ith training examples
   h:hypothesis(假设函数)
   
   线性回归（linear regression）,单变量线性回归（univeriate linear regression–single variable）
4. 代价函数（cost function）：